• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 2 answers
  • Subscribers 46 subscribers
  • Views 1295 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

ADS1220 Drift with time

Kevin Hughes
Intellectual 300 points

Other Parts Discussed in Thread: ADS1220

Hello,

The ADS1220 datasheet shows a drift specification for the internal voltage reference, but does not state what the expected drift of the actual measurements is in various configurations.  We are using an external reference in a ratiometric configuration, and need to estimate what the accuracy drift over days, months, years, will be for the ADC measurements.  

Does such a specification exist?  Is there any data TI can share on this part?

Thank you,

Kevin

  • 0
    TI__Guru** 113765 points

    Hi Kevin,

    If you are using an external reference in a ratiometric configuration, then the reference drift will cancel out in the result.  The effects of temperature drift will remain for the ADC with respect to gain and offset, so these are the key parameters to consider with this type of measurement.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Intellectual 300 points
    Hi Bob,
    Thanks for the reply. So, if the part only operated at a constant temperature, there would be no drift with time in the measurement? (I am ignoring the physical change in the sensor here, just looking at the ADC measurement itself).
  • 0 in reply to Kevin Hughes
    TI__Guru** 113765 points

    Hi Kevin,

    Yes that would be correct within the level of noise of the converter.  However, there is always some heating from self-heating and from the heating of devices around the ADS1220.  There is one other caveat I should mention and that is with respect to the stability of the analog supply.  If the analog supply voltage drifts too much  or if the supply is very noisy you will see some degradation of the result relative to rated PSR for the device.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Intellectual 300 points
    Hello,
    Apologies for digging this question up, but there's one followup I'm not clear on:
    The ADS1220 datasheet lists an offset voltage of +/-30uV max for G=1 at 25C, and then states a max drift of 0.3uV/C. I understand the effect of temperature drift, but could you please clarify whether the baseline offset voltage will remain static across time?

    If, for example, we calibrate the offset voltage at the factory, and a customer uses the device at the same temperature as it was calibrated, will the offset voltage be the same? Clearly not "exactly" the same, but approximately what level of aging in offset voltage is expected?

    Thermal hysteresis presumably will play a role in this as well - do you have specifications or measurement for that parameter?


    Thank you very much for your support,
    Kevin
  • 0 in reply to Kevin Hughes
    TI__Guru** 113765 points

    Hi Kevin,

    We don't characterize long term aging of the offset on our devices.  As the input is chopped to reduce the offset, there really isn't much that would affect the offset over time.  So the biggest difference here would be noise relative to your initial calibration.

    With regard to offset, you could use the internal short feature within the mux of the ADS1220 which would give you a code result that could be subtracted from your measurement at each power up.  Obviously that would not account for total system offset, but if that is of large concern I don't think the stability of the ADC would be your only problem.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Intellectual 300 points
    Hi Bob,
    Thank you for the helpful reply - good reminder that the voltage offset takes into account the chopping action.

    You are correct, stability of the ADC is not our only concern. Since these voltage levels do matter, could you list the other potential drift sources in the ADC? I know the answer may be "read the datasheet," but I want to make sure we're not missing something that would be obvious to you in determining the ADC contribution to total system uncertainty.

    Thank you,
    Kevin
  • 0 in reply to Kevin Hughes
    TI__Guru** 113765 points

    Hi Kevin,

    I believed I answered this in my initial response.  Most likely gain error will be the largest contributor.  This will depend on the stability of the reference source and how you are establishing the ratiometric measurement.

    Best regards,

    Bob B